1 of 20

Short Pulse High Gradient Accelerating Structures

Sergey V. Kuzikov

Euclid TechLabs, LLC, Bolingbrook, IL

Outline:

  1. Short pulse high-gradient acceleration concept:
    1. Breakdown threshold
    2. Efficiency
    3. Lifetime
  2. Experiments with nanosecond X-band components;
  3. THz single-cycle structures;
  4. Mm- and cm- single-cycle structures:

a) GV/m wakefield structures

b) multi-mode wakefield structures.

2 of 20

2

Euclid Techlabs/Euclid Beamlabs

Euclid Techlabs, LLC is a research and development company specializing in linear particle accelerators, ultrafast electron microscopy, and advanced material technologies. The company was formed in 2003. Euclid Beamlabs LLC, formed in the winter of 2014, is a sister (spin-off) company of Euclid Techlabs LLC, particularly to commercialize industrial accelerator and related advanced material technologies developed at Euclid Techlabs. Euclid has developed expertise and products in several innovative technologies: time-resolved ultra-fast electron microscopy; ultra-compact linear accelerators; electron guns with thermionic, field emission or photo-emission cathodes; fast tuners for SRF cavities; advanced dielectric materials; HPHT and CVD diamond growth and applications; thin-film for accelerator technologies; Present: 27 people research staff (researchers, engineers, technicians) and 5 administrative. 16 PhDs in accelerator physics and material science, 32 staff. 2 labs: Bolingbrook, IL (accelerator R&D lab) and Beltsville, MD (material science lab). Long term collaborations with National Labs and Institutes: ANL, Fermilab, BNL, Jlab, LBL, SLAC, LANL, NIST, NIU, IIT, etc.

www.euclidtechlabs.com

Fermilab

Euclid

Argonne

Euclid

UMD

NRL, JLab

NIST

BNL

3 of 20

3

Products & Capabilities Snapshot

Products

  • UltraFast Pulser (UFPTM) for TEM
  • Dislocation free diamond for Xray optics
  • Compact X-Ray Source
  • NCRF and SRF electron sources
  • Low loss ceramics (linear and non-linear)
  • LINAC
  • RF window
  • In flange BPM

Capabilities

  • Femtosecond Laser Ablation System
  • Thin Film Deposition Lab
  • EM Testing Lab
  • Radiation Shielding/Testing Lab

4 of 20

  • Feeding by short, high repetition rate pulses, in order to avoid breakdown and excessive pulse heating according to scaling laws:
  • Accelerating fields more than 100 MV/m have already been obtained at ~100 ns pulses in X-band. Extrapolating scaling law one must conclude that in order to reach 1 GV/m, as short as ~ 1 ps (1 THz width) pulses are needed.

Challenges:

  • Broad band systems are necessary;
  • High repetition rate;
  • High efficiency;
  • Structure lifetime must be large enough.

Pulse Length Scaling & Challenges

5 of 20

High-Gradient X-Band Gun Fed by 9 ns RF Pulses

11.7 GHz field structure

RF

6 of 20

6

High-Gradient Test and Beam Energy Characterization

~3% fluctuation

1000 shots as total

  • Energy measured by the spectrometer dipole
  • ~3% fluctuation, likely due to the drive charge instability.
  • Max achieved gradient is 388 MV/m from the beam energy measurement

Simulated kinetic-energy isoclines as function of RF gun operating conditions K(E0, φ0) and retrieved operating points (“+” symbols). The shaded areas represent the uncertainty on the measured φ0 and inferred E0 values.

 

BDR~10-6 , 100 pC

Low average dark current

7 of 20

7

1. A. D. Cahill et al. High gradient experiments with X-band cryogenic copper accelerating cavities, 2018.

Short Pulse Structures at Cryogenic Temperatures

Breakdown rate vs gradient [1]

For the 11.7 GHz 9 ns gun the cathode gradient at 45 K could be as high as more than 500 MV/m, maximum surface electric field might be as high as 750 MV/m.

2. V.A. Dolgashev, Recent High Gradient Tests at SLAC, 2016.

Peak pulse heating is a reasonably good predictor the breakdown probability [2].

8 of 20

8

A Nanosecond Accelerating Structure with Individually Fed Cells

5-cell side coupled accelerating structure: E-field structure of the operating mode at 11.7 GHz (a), several structures assembly (b), S11 for the feeding antenna (c).

splitter

Bent delay waveguides

5 independent cells

Power

150 MW

RF pulse length

10 ns

Q-factor

350

Accelerator type

π-mode

Bending angle

69.2°

Beam channel

∅10 mm

Shunt impedance

44 MΩ/m

Surface Emax

750 MV/m

Surface Bmax

0.8 MA/m

Parameters of the 11.7 GHz, 300 MV/m, 5-cell structure (T=300 K).

The TM010 accelerating cells are powered individually!

This principle allows:

- to mitigate the influence of RF breakdown if appears in one cell;

- to increase the shunt impedance due to the smaller than usual beam channel;

- to reduce sensitivity to the tolerances on the size of the cells.

a)

b)

c)

9 of 20

9

Because the luminosity is required to be high, the repetition rate for a short-pulse structure must be increased by factor of pulse length reduction. It could be as high as sub-MHz level.

Estimations for Structure Lifetime

long-pulse structure

short-pulse

structure

τ0

τ

τ=τ0/N

N

1

2

3

 

 

for N short pulses, i.e.

 

Let us take BDRN=BDR and assume that in case of breakdown event all RF power stored in a cell can be deposited in material erosion:

 

To “kill” the structure completely one need evaporating as much material as:

 

Eventually, the full number of macro- pulses to destroy the structure:

micro-pulses

macro-pulse

 

for a long-pulse structure.

10 of 20

10

Frequency

 10 GHz

 10 GHz

Waveguide type

WR90

WR90

RF pulse duration

1 ns

0.5 ns

RF pulse period

10 ns

5 ns

Peak RF power

170 MW

140 MW

Power gain

47

55

Bunch charge

1 nC

3.5 nC

Efficiency

0.7%

6%

High-Efficiency Short Pulse Acceleration

Steady state traveling RF pulse (blue) and the feeding RF pulse (red).

Resonator with equidistant mode spectrum.

TE10

time

Eacc

Field in accelerating gap

500 MV/m accelerating structure

Delay waveguide

Fourier spectrum of pulses and eigen mode spectrum of resonator

Power gain (blue) and field enhancement in accelerating gap (red)

11 of 20

11

Beam

channel

E

E

TEM travelling wave delay lines

Coaxial Travelling Wave Structure

Frequency dispersion is almost absent!

Power gain (~20) depends on losses in delay waveguides,

It can be increased using special coating or working at cryogenic temperatures.

E-field

12 of 20

12

High-Power Short-Pulse RF Sources

  1. G. G. Denisov, S. V. Kuzikov, A. V. Savilov. Q-switching in the electron backward-wave oscillator, Physics of plasmas 18, 103102 (2011).
  2. S.V. Kuzikov, A.V. Savilov, Parametric Phase Locking in an Electron RF Oscillator, Phys. Rev. Lett. 110, 174801 (2013).

BWO with Q-switching:

- Stationary pulse-periodic generation of short rf pulses.

- RF pulses are phase locked by Q-modulation shape.

Q-switched BWO : Efficiency = 60-70%

peak rf power = 350 %

13 of 20

13

VBS test stand with installed test resonator: a – overall view, b – test resonator in vacuum chamber installed in the bottom, c – view through quartz window to YAG-screen.

VBS beam modulation (5 Hz)

Electron beam

Beam on

(50% of time)

Beam off

(50% of time)

RF Modulation by Electron Beam and AC Bias Voltage

Diamond

Bias (3 kV) modulation frequency is 100 Hz.

(I=1-7 μA)

Avalanche effect

14 of 20

14

Emitted Terahertz spectrum

*C. Vicario et al. Optics Letters, Vol. 39, Iss. 23, 2014.

High-fields have been produced by optical rectification in a large-size organic crystal by a powerful pump lasing. The phase-locked single-cycle pulses carry peak fields of more than 80 MV/cm at diffraction limited spot size. The scheme has 3% conversion efficiency.

Generation of mJ Single THz Pulses with Electric Field Exceeding 80 MV/cm*

15 of 20

15

E-field distributions at the lens while focusing the short THz pulse, for the time correspondent to THz pulse arrives to lens (a), for time when focusing begins (b), for time when focusing is close to maximum (c), and in maximum of focusing (d).

Multi-Cell Accelerating Structures Driven by Focused THz Pulses

Timing of THz pulses in the structure.

16 of 20

16

High current drive beam

Witness beam

Single Cycle Wakefield Accelerating Structure

17 of 20

A Wakefield Single-Cycle Accelerating Structure

Excitation of a single-cycle wake in a gap with a parabolic shape. Time flows evenly as a-b-c-d.

a)

b)

c)

d)

Single-cycle acceleration cell.

Shape of the single-cycle wake (a) and spectrum of this wake (b) at the output of the parabolic mirror with a length of 9.6 mm and an aperture of 150 mm generated by a 50-nC, 5-mm-long (Gaussian) single bunch.

a)

b)

18 of 20

Field excitation by a single bunch in parabolic mirrors, and evolution of this field in the delay line waveguides. The bunch travels from left to right.

Summation of pulses in the combiner and focusing of the resulted single-cycle pulse by the parabolic mirror. Time flows evenly from left to right.

Field shapes for the pulse entering the pulse combiner (green curve), for the pulse at the entrance of the focusing parabola (pink curve), and at the focus of the final parabolic mirror for 50 nC.

Parameters of the accelerating structures.

Parameters

 

 

 

Bunch charge, nC

50

150

250

Number of bunches

5

2

1

Number of cells

11

20

11

Accelerating field, MV/m

170

470

570

19 of 20

19

A Multi-Mode Wakefield Accelerating Structure

Field patterns at the time points shown in Fig. 5: a – point A, b – point B, c – point C, d – point D.

E-field mode structures in a slice of the square resonator: a – TM1,1,0 , b – TM3,3,0, c - TM5,5,0, d- TM7,7,0, e – TM9,9,0.

E-field in the middle of the resonator with mode spectrum correction excited by 100 bunches.

Enlarged portion of the previous figure.

S.V. Kuzikov et al, Phys. Rev. Lett. 104, 214801 (2010). S.V. Kuzikov et al. Phys. Rev. Lett. ST Accel. Beams, Vol. 13, No.7, 071303 (2010).

20 of 20

Conclusion

  1. Short-pulse structures show prospects to achieve ~GV/m gradients. Its can have high efficiencies that are comparable with efficiencies of classical structures.
  2. Breakdown threshold (theoretical and experimental) study for ultra short RF pulses is needed. The temperature dependence is also extremely interesting.
  3. High repetition rate, short pulse RF sources are necessary. So far experiments with wakefield class structures are only available.
  4. The synergistic effect of short pulse technology and cryogenic technology would allow reaching world record gradients as well as high efficiency due to a higher wall conductance.